Project Abstract PDA is the most common neoplasm of the pancreas, and is soon to be the second most common cause of cancer deaths in the United States. Surgical resection in Stage I/II patients provides the only opportunity for cure, yet >80% of patients will recur and die of their disease within 2-3 years2. The statistics for Stage III and Stage IV PDA are more dismal, having 12 and 6 month median overall survival times, respectively. Outside of BRCA2 mutations that confer sensitivity to platinum salts or PARP inhibition, or immune checkpoint inhibitors in patients with mismatch repair deficiency, there are few actionable targets in the PDA genome. Thus, it is essential that novel strategies are developed to extend survival. One innovative way to do so is to ?treat evolution with evolution?. However, it is first imperative that we develop a deep understanding of PDA evolutionary biology. Our efforts will be focused on three questions with clear mechanistic and translational relevance to this ultimate goal. First, what are the features of clinically relevant intratumoral heterogeneity at the genetic and transcriptional level? Second, how do cell autonomous and non-cell autonomous factors influence the evolutionary dynamics of PDA? Third, how do PDA therapies influence evolutionary trajectories, and can they be more effectively used within the evolutionary context of a tumor? We will rely on whole exome or whole genome sequenced samples of primary and metastatic pancreatic cancer tissues, single cell technologies for copy number alterations or RNA expression, long-term evolution experiments, mouse models and computational models to address these questions. We aim to use the information gained over the period of this work to develop metrics of heterogeneity that will inform clinical management, including identification of the optimal agents and timing of administration based on the evolutionary context of the patients' PDA. Such questions are of broad interest in cancer biology in general and have a strong likelihood to impact upon other tumor types as well.